Helen Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Helen Xu has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Helena R. Ma
Department of Biomedical Engineering, Duke University, Durham, USA
2 shared publications - Lingchong You
Department of Biomedical Engineering, Duke University
2 shared publications Deverick J. Anderson
Division of Infectious Diseases, Department of Medicine, Duke University School of Medicine, Durham, USA
1 shared publication- Gowri Nayar
Stanford University
1 shared publication - Issah Samori
Department of Bioengineering, Stanford University, Stanford, CA, USA
1 shared publication - Kyeri Kim
Department of Biomedical Engineering, Duke University, Durham, USA
1 shared publication
Latest Publications
- Open Access
Homology-Based Variant-Effect Predictors Break Down on Cytochrome P450 Pharmacogenes *
Cytochrome P450 (CYP) enzymes metabolize roughly three-quarters of clinically used drugs, and genetic variation in these enzymes is a leading source of interindividual differences in drug response. Predicting a variant’s functional effect is therefore critical, yet the consequences of most CYP variants remain unknown. Many state-of-the-art variant-effect predictors rest on a homology-based paradigm that scores variants by evolutionary conservation — an assumption that pharmacogenes, including CYPs, violate. Focusing on human CYPs, we show that homology-based models fail systematically. AlphaMissense (AM) assigns variants to its “ambiguous” class at nearly twice the proteome-wide rate across six clinically important CYPs, and within that class its scores are essentially uninformative about measured CYP2C9 deep mutational scan (DMS) activity ( ρ = −0.130). Using catalytic activity and cellular abundance from the same CYP2C9 DMS, we show that AM is largely an abundance signal: it tracks activity well for destabilizing, low-abundance variants ( ρ = −0.568) but far more weakly for abundance-independent variants ( ρ = −0.357; Fisher r-to-z z = −8.99, p < 10 −18 ). We further hypothesized that non-homology-based features (sequence position, substitution chemistry, distance to the substrate-recognition sites, and secondary structure) might account for where AM errs, but their explanatory power is weak. Using CYP2C9 DMS activity as ground truth, we built a k -nearest-neighbors model over ESM-2 embeddings and ensembled it with AM and ESM-2 masked marginal probability, raising the correlation between predicted and measured activity within the ambiguous class from ρ = 0.080 for AM alone to 0.399, and reaching an overall ρ = 0.721 that exceeds every out-of-the-box predictor we tested ( ρ = 0.523 to 0.686). However, neither this improved ensemble model nor any other method agrees significantly with clinical annotations. Drawing on evidence that a variant’s effect can depend on the drug, we hypothesize that substrate identity is the key missing feature in current models, and that predicting function for these multi-substrate enzymes may require redefining function as substrate-conditioned.
Open AccessHomology-Based Variant-Effect Predictors Break Down on Cytochrome P450 Pharmacogenes *
Cytochrome P450 (CYP) enzymes metabolize roughly three-quarters of clinically used drugs, and genetic variation in these enzymes is a leading source of interindividual differences in drug response. Predicting a variant’s functional effect is therefore critical, yet the consequences of most CYP variants remain unknown. Many state-of-the-art variant-effect predictors rest on a homology-based paradigm that scores variants by evolutionary conservation — an assumption that pharmacogenes, including CYPs, violate. Focusing on human CYPs, we show that homology-based models fail systematically. AlphaMissense (AM) assigns variants to its “ambiguous” class at nearly twice the proteome-wide rate across six clinically important CYPs, and within that class its scores are essentially uninformative about measured CYP2C9 deep mutational scan (DMS) activity ( ρ = −0.130). Using catalytic activity and cellular abundance from the same CYP2C9 DMS, we show that AM is largely an abundance signal: it tracks activity well for destabilizing, low-abundance variants ( ρ = −0.568) but far more weakly for abundance-independent variants ( ρ = −0.357; Fisher r-to-z z = −8.99, p < 10 −18 ). We further hypothesized that non-homology-based features (sequence position, substitution chemistry, distance to the substrate-recognition sites, and secondary structure) might account for where AM errs, but their explanatory power is weak. Using CYP2C9 DMS activity as ground truth, we built a k -nearest-neighbors model over ESM-2 embeddings and ensembled it with AM and ESM-2 masked marginal probability, raising the correlation between predicted and measured activity within the ambiguous class from ρ = 0.080 for AM alone to 0.399, and reaching an overall ρ = 0.721 that exceeds every out-of-the-box predictor we tested ( ρ = 0.523 to 0.686). However, neither this improved ensemble model nor any other method agrees significantly with clinical annotations. Drawing on evidence that a variant’s effect can depend on the drug, we hypothesize that substrate identity is the key missing feature in current models, and that predicting function for these multi-substrate enzymes may require redefining function as substrate-conditioned.
- Open Access
- 27 September 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-14
Private benefit of β-lactamase dictates selection dynamics of combination antibiotic treatment
β-lactam antibiotics have been prescribed for most bacterial infections since their discovery. However, resistance to β-lactams, mediated by β-lactamase (Bla) enzymes such as extended spectrum β-lactamases (ESBLs), has become widespread. Bla inhibitors can restore the efficacy of β-lactams against resistant bacteria, an approach which preserves existing antibiotics despite declining industry investment. However, the effects of combination treatment on selection for β-lactam resistance are not well understood. Bla production confers both private benefits for resistant cells and public benefits which faster-growing sensitive cells can also exploit. These benefits may be differentially impacted by Bla inhibitors, leading to non-intuitive selection dynamics. In this study, we demonstrate strain-to-strain variation in effective combination doses, with complex growth dynamics in mixed populations. Using modeling, we derive a criterion for the selection outcome of combination treatment, dependent on the burden and effective private benefit of Bla production. We then use engineered strains and natural isolates to show that strong private benefits of Bla are associated with increased selection for resistance. Finally, we demonstrate that this parameter can be coarsely estimated using high-throughput phenotyping of clonal populations. Our analysis shows that quantifying the phenotypic responses of bacteria to combination treatment can facilitate resistance-minimizing optimization of treatment.
- 27 September 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-14
Open AccessPrivate benefit of β-lactamase dictates selection dynamics of combination antibiotic treatment
β-lactam antibiotics have been prescribed for most bacterial infections since their discovery. However, resistance to β-lactams, mediated by β-lactamase (Bla) enzymes such as extended spectrum β-lactamases (ESBLs), has become widespread. Bla inhibitors can restore the efficacy of β-lactams against resistant bacteria, an approach which preserves existing antibiotics despite declining industry investment. However, the effects of combination treatment on selection for β-lactam resistance are not well understood. Bla production confers both private benefits for resistant cells and public benefits which faster-growing sensitive cells can also exploit. These benefits may be differentially impacted by Bla inhibitors, leading to non-intuitive selection dynamics. In this study, we demonstrate strain-to-strain variation in effective combination doses, with complex growth dynamics in mixed populations. Using modeling, we derive a criterion for the selection outcome of combination treatment, dependent on the burden and effective private benefit of Bla production. We then use engineered strains and natural isolates to show that strong private benefits of Bla are associated with increased selection for resistance. Finally, we demonstrate that this parameter can be coarsely estimated using high-throughput phenotyping of clonal populations. Our analysis shows that quantifying the phenotypic responses of bacteria to combination treatment can facilitate resistance-minimizing optimization of treatment.
- Open Access
- 1 December 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-17
Autoencoder neural networks enable low dimensional structure analyses of microbial growth dynamics
The ability to effectively represent microbiome dynamics is a crucial challenge in their quantitative analysis and engineering. By using autoencoder neural networks, we show that microbial growth dynamics can be compressed into low-dimensional representations and reconstructed with high fidelity. These low-dimensional embeddings are just as effective, if not better, than raw data for tasks such as identifying bacterial strains, predicting traits like antibiotic resistance, and predicting community dynamics. Additionally, we demonstrate that essential dynamical information of these systems can be captured using far fewer variables than traditional mechanistic models. Our work suggests that machine learning can enable the creation of concise representations of high-dimensional microbiome dynamics to facilitate data analysis and gain new biological insights.
- 1 December 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-17
Open AccessAutoencoder neural networks enable low dimensional structure analyses of microbial growth dynamics
The ability to effectively represent microbiome dynamics is a crucial challenge in their quantitative analysis and engineering. By using autoencoder neural networks, we show that microbial growth dynamics can be compressed into low-dimensional representations and reconstructed with high fidelity. These low-dimensional embeddings are just as effective, if not better, than raw data for tasks such as identifying bacterial strains, predicting traits like antibiotic resistance, and predicting community dynamics. Additionally, we demonstrate that essential dynamical information of these systems can be captured using far fewer variables than traditional mechanistic models. Our work suggests that machine learning can enable the creation of concise representations of high-dimensional microbiome dynamics to facilitate data analysis and gain new biological insights.